Help Center/ Data Replication Service/ FAQs/ Real-Time Synchronization/ Configuring Remote Connection to a GaussDB Database/ Configuration Method for a Synchronization Task from GaussDB Distributed in Single-Task Mode
Updated on 2025-02-17 GMT+08:00

Configuration Method for a Synchronization Task from GaussDB Distributed in Single-Task Mode

Method 1: Modify the configuration file.

Procedure

  1. Log in to all nodes of the distributed GaussDB instance.
  2. Query the location of the configuration file.

    find / -name "pg_hba.conf"
    Or
    find / -name "gs_hba.conf"

  3. Open the pg_hba.conf file. Retain sha256 and insert trust of the specified single-mode task user in the line above sha256.

    • CN configuration
      host replication <user> 0.0.0.0/0 sha256
      Figure 1 CN configuration
    • DN configuration

      host replication <user> 0.0.0.0/0 trust

      Figure 2 DN configuration

      For the DN configuration, do not set <User> to all. Otherwise, connections to DNs on a distributed source instance in multi-task mode will be affected.

  4. Run the following statement as a system administrator on the source database or restart the DB instance to apply the change:

    select pg_reload_conf();

Method 2: Use gs_guc.

  1. For a cloud database, run the following command to access the sandbox. For a self-built database, run the source gauss_env_file command to configure the environment variable file.

    Log in to any database node, switch to user Ruby, and run the su - Ruby command.

    1. Versions earlier than 2.0 (V500R002C00)
      /usr/sbin/chroot /var/chroot /bin/bash
      source /etc/profile && source ~/.bashrc
    2. 2.0 (V500R002C00) and later versions
      chroot /var/chroot
      source /etc/profile; source /home/Ruby/.bashrc ;source /home/Ruby/gauss_env_file

    To change a GaussDB instance on Huawei Cloud, you need to contact the database on-call personnel.

  2. If only the single-task mode is used, the drs1 user is used as an example.

    gs_guc reload -Z coordinator -N all -I all -h "host all drs1 0.0.0.0/0 sha256"

    gs_guc reload -Z coordinator -N all -I all -h "host replication drs1 0.0.0.0/0 sha256"

    Configure the replication connection between the CN and DN. The following uses network segment 192.168.0.0/16 of the CN as an example.

    gs_guc reload -Z datanode -N all -I all -h 'host replication drs1 192.168.0.0/16 trust'

    If only the multi-task mode is used, the drs2 user is used as an example.

    gs_guc reload -Z datanode -Z coordinator -N all -I all -h "host all drs2 0.0.0.0/0 sha256"

    gs_guc reload -Z datanode -N all -I all -h 'host replication drs2 0.0.0.0/0 sha256'

    If you use both the multi-task and single-task modes for the same source instance, you are advised to plan migration users and configure the task as required.